Research On The Method Of Building Many-objective Optimization And Decision Support In The Concept Design Stage

The building industry is changing from high energy consuming to high performance.Building energy efficiency along with other performance needs to be optimized together,which can be achieved by multi-objective optimization.In building concept design stage,many-objective optimization problems with more than three objectives are often involved,and it is necessary to integrate many-objective optimization into this stage where design decisions greatly impact the building performance.However,the traditional building performance simulation based multiobjective optimization is inefficient,and unable to solve many-objective problems.In addition,the existing building multi-objective optimization methods provide insufficient decision support,and there is a lack of systematic tools and processes.To solve the above problems,this paper first combed the evolution of the building performance optimization design,analyzed the building multi-objective optimization design theories,and summarized the bottlenecks of the existing methods as well as the key points for making high performance oriented design decisions.A method of building many-objective optimization and decision support in the concept design stage was proposed.By developing a building performance prediction based many-objective optimization process and an SOM neural network based decision support process,the limitation on the number of objectives was broken with an improvement on the optimization search speed,and the decision support problem was solved.An optimization platform integrated into a parametric modeling software was created and a decision support tool was developed,which promoted the application of the proposed method.Finally,this method was verified by the case study of a public building.Building Energy Use Intensity(EUI),spatial Daylight Autonomy(s DA),Useful Daylight Illuminance(UDI)and Building Envelope Cost(BEC)performance of the building was optimized,and the final building design decision was determined.The case study results proved that the proposed method can improve the optimization efficiency and greatly reduce the difficulty of decision-making,also promote many building performance simultaneously in the concept design stage.In addition,this method provided the platform and tool that can be used in the building concept design stage by the architects,which can promote the wide application of the building multi-objective optimization,and it is of great significance for the development of high-performance buildings in the new era.